Signal Processing
Automated Tinnitus Detection Through Dual-Modality Neuroimaging: EEG Microstate Analysis and Resting-State fMRI Classification Using Deep Learning
Kiana Kiashemshaki, Sina Samieirad, Sarvenaz Erfani et al. (+3 more)
Objective: Tinnitus affects 10-15% of the population yet lacks objective diagnostic biomarkers. This study applied machine learning to
EEG and fMRI data to identify neural signatures distinguishing tinnitus patients from healthy controls. Methods: Two
datasets were analyzed: 64-channel
EEG recordings from 80 participants (40 tinnitus, 40 controls) and resting-state fMRI data from 38 participants (19 tinnitus, 19 controls).
EEG analysis extracted microstate features across four to seven clustering states and five frequency bands, producing 440 features per subject. Global Field Power signals were also transformed into wavelet images for deep learning. fMRI data were analyzed using slice-wise convolutional neural networks and hybrid models combining pre-trained architectures (VGG16, ResNet50) with Decision Tree, Random Forest, and SVM classifiers. Model performance was evaluated using 5-fold cross-validation based on accuracy, precision, recall, F1-score, and ROC-AUC. Results:
EEG microstate analysis revealed altered network dynamics in tinnitus, particularly reduced gamma-band microstate B occurrence (healthy: 56.56 vs tinnitus: 43.81, p < 0.001) and diminished alpha coverage. Tree-based classifiers achieved up to 98.8% accuracy, while VGG16 on wavelet-transformed
EEG yielded 95.4% and 94.1% accuracy for delta and alpha bands, respectively. fMRI analysis identified 12 high-performing axial slices (>=90% accuracy), with slice 17 reaching 99.0%. The hybrid VGG16-Decision Tree model achieved 98.95% +/- 2.94% accuracy. Conclusion:
EEG and fMRI provided effective neural biomarkers for tinnitus classification. Tree-based and hybrid models demonstrated superior performance, highlighting tinnitus as a multi-network disorder requiring
multimodal analysis.
Read the full abstract ▾
All authors Kiana Kiashemshaki, Sina Samieirad, Sarvenaz Erfani, Aryan Jalaeianbanayan, Nasibeh Asadi Isakan, Hossein Najafzadeh
Systems and Control
Excitation of Looped Bistable Bands for High-Speed Linear Actuation
Sareum Kim and Josie Hughes
Soft robotics increasingly relies on smart materials and innovative structures, with bistable tape springs emerging as a promising option. These structures exhibit intriguing dynamic behaviors, such as oscillation, due to their inherent bistability. This paper explores the high-speed linear amplification of motion achieved through the excitation of a looped bistable tape spring. When looped, the tape spring forms two distinct joints, facilitating smooth oscillation. Mounted on a linear guide and driven by a crank mechanism with varying frequency, the system converts input oscillations into amplified linear motion at resonance. This study highlights the potential of bistable tape springs high speed reciprocating linear motion.
Read the full abstract ▾
Audio and Speech Processing
VAPO: End-to-end Slide-Enhanced Speech Recognition with Omni-modal Large Language Models
Rui Hu, Delai Qiu, Yining Wang et al. (+2 more)
Omni-modal large
language models (OLLMs) offer a promising end-to-end solution for slide-enhanced
speech recognition due to their inherent
multimodal capabilities. However, we found a fundamental issue faced by OLLMs: Visual Interference, where models show a bias towards visible text over auditory signals, causing them to hallucinate slide content that was never spoken. To address this, we propose Visually-Anchored Policy Optimization (VAPO), which aims to reshape models' inference process to follow the human-like “Look-then-Listen” inference chain. Specifically, we design a temporally decoupled policy: the model first extracts visual priors in a <think> block to serve as semantic anchors, then generates the transcription in an <answer> block. The policy is optimized via multi-objective reinforcement learning. Furthermore, we introduce SlideASR-Bench, a comprehensive
benchmark designed to address the scarcity of entity-rich data, comprising a large-scale synthetic corpus for training and a challenging real-world test set for evaluation. We conduct extensive evaluations demonstrating that VAPO effectively eliminates visual interference and achieves state-of-the-art performance on SlideASR-Bench and public
datasets, significantly reducing entity recognition errors in specialized domains.
Read the full abstract ▾
All authors Rui Hu, Delai Qiu, Yining Wang, Shengping Liu, Jitao Sang
Systems and Control · Cross-listed from cs.RO
GATO: GPU-Accelerated and Batched Trajectory Optimization for Scalable Edge Model Predictive Control
Alexander Du, Emre Adabag, Gabriel Bravo-Palacios et al. (+1 more)
While Model Predictive Control (MPC) delivers strong performance across robotics applications, solving the underlying (batches of) nonlinear trajectory optimization (TO) problems online remains computationally demanding. Existing GPU-accelerated approaches either parallelize single solves, handle large batches at sub-real-time rates, or sacrifice model generality for speed. This leaves a large gap in solver performance for many state-of-the-art MPC applications that require real-time batches of tens to low-hundreds of solves. As such, we present GATO, an open source, GPU-accelerated, batched TO solver co-designed across
algorithm, software, and computational hardware to deliver real-time throughput for these moderate batch size regimes. Our approach leverages a combination of block-, warp-, and thread-level parallelism within and across solves for ultra-high performance. We demonstrate the effectiveness of our approach through a combination of: simulated
benchmarks showing speedups of 18-21x over CPU baselines and 1.4-16x over GPU baselines as batch size increases; case studies highlighting improved disturbance rejection and convergence behavior; and finally a validation on hardware using an industrial manipulator. We open source GATO to support reproducibility and adoption.
Read the full abstract ▾
All authors Alexander Du, Emre Adabag, Gabriel Bravo-Palacios, Brian Plancher
Signal Processing · Cross-listed from cs.LG
Transformer-Based Indirect Structural Health Monitoring of Rail Infrastructure with Attention-Driven Detection and Localization of Transient Defects
Sizhe Ma, Katherine A. Flanigan, Mario Bergés et al. (+1 more)
Indirect structural health monitoring (iSHM) for broken rail detection using onboard sensors presents a cost-effective paradigm for railway track assessment, yet reliably detecting small, transient anomalies (2-10 cm) remains a significant challenge due to complex vehicle dynamics, signal noise, and the scarcity of labeled data limiting supervised approaches. This study addresses these issues through unsupervised deep learning. We introduce an incremental synthetic data
benchmark designed to systematically evaluate model robustness against progressively complex challenges like speed variations, multi-channel inputs, and realistic noise patterns encountered in iSHM. Using this
benchmark, we evaluate several established unsupervised models alongside our proposed
Attention-Focused
Transformer. Our model employs a self-
attention mechanism, trained via reconstruction but innovatively deriving anomaly scores primarily from deviations in learned
attention weights, aiming for both effectiveness and computational efficiency. Benchmarking results reveal that while
transformer-based models generally outperform others, all tested models exhibit significant vulnerability to high-frequency localized noise, identifying this as a critical bottleneck for practical deployment. Notably, our proposed model achieves accuracy comparable to the state-of-the-art solution while demonstrating better inference speed. This highlights the crucial need for enhanced noise robustness in future iSHM models and positions our more efficient
attention-based approach as a promising foundation for developing practical onboard anomaly detection systems.
Read the full abstract ▾
All authors Sizhe Ma, Katherine A. Flanigan, Mario Bergés, James D. Brooks
Audio and Speech Processing
SALAD-VAE: Semantic Audio Compression with Language-Audio Distillation
Sebastian Braun, Hannes Gamper and Dimitra Emmanouilidou
Modern generative and
multimodal models increasingly rely on compact latent representations that trade and balance semantic richness with high-fidelity reconstruction. We introduce SALAD-VAE, a continuous and highly compact semantic Audio Variational Autoencoder, which operates in the frequency domain and achieves state-of-the-art compression with very low latent frame rate (7.8 Hz) while surfacing semantic structure and producing high audio quality. We enhance the standard VAE semantic losses and augmentation, specifically contrastive learning and CLAP-based
embedding distillation, enabling it to generalize across diverse audio domains. With a significantly less computational complex architecture than comparable state-of-the-art VAEs, SALAD-VAE matches their reconstruction quality while it consistently outperforms them on a wide range of classification
benchmarks. Furthermore, the proposed additional loss function provides a trained CLAP projection layer, which can be used zero-shot audio captioning and classification matching pretrained CLAP audio-text
embeddings.
Read the full abstract ▾
Systems and Control · Cross-listed from cs.LG
Accuracy, Memory Efficiency and Generalization: A Comparative Study on Liquid Neural Networks and Recurrent Neural Networks
Shilong Zong, Alex Bierly, Almuatazbellah Boker et al. (+1 more)
This review aims to conduct a comparative analysis of
liquid neural networks (LNNs) and traditional recurrent neural networks (RNNs) and their variants, such as long short-term
memory networks (LSTMs) and gated recurrent units (GRUs). The core dimensions of the analysis include model accuracy,
memory efficiency, and generalization ability. By systematically reviewing existing research, this paper explores the basic principles, mathematical models, key characteristics, and inherent challenges of these neural network architectures in processing sequential data. Research findings reveal that LNN, as an emerging, biologically inspired, continuous-time dynamic neural network, demonstrates significant potential in handling noisy, non-stationary data, and achieving out-of-distribution (OOD) generalization. Additionally, some LNN variants outperform traditional RNN in terms of parameter efficiency and computational speed. However, RNN remains a cornerstone in sequence modeling due to its mature ecosystem and successful applications across various tasks. This review identifies the commonalities and differences between LNNs and RNNs, summarizes their respective shortcomings and challenges, and points out valuable directions for future research, particularly emphasizing the importance of improving the scalability of LNNs to promote their application in broader and more complex scenarios.
Read the full abstract ▾
All authors Shilong Zong, Alex Bierly, Almuatazbellah Boker, Hoda Eldardiry
Signal Processing · Cross-listed from cs.LG
Estimating Fair Graphs from Graph-Stationary Data
Madeline Navarro, Andrei Buciulea, Samuel Rey et al. (+2 more)
We estimate fair graphs from graph-stationary nodal observations such that connections are not biased with respect to sensitive attributes. Edges in real-world graphs often exhibit preferences for connecting certain pairs of groups. Biased connections can not only exacerbate but even induce unfair treatment for downstream graph-based tasks. We therefore consider group and individual
fairness for graphs corresponding to group- and node-level definitions, respectively. To evaluate the
fairness of a given graph, we provide multiple bias metrics, including novel measurements in the spectral domain. Furthermore, we propose Fair Spectral Templates (FairSpecTemp), an optimization-based method with two variants for estimating fair graphs from stationary graph signals, a general model for graph data subsuming many existing ones. One variant of FairSpecTemp exploits commutativity properties of graph stationarity while directly constraining bias, while the other implicitly encourages fair estimates by restricting bias in the graph spectrum and is thus more flexible. Our methods enjoy high probability performance bounds, yielding a conditional tradeoff between
fairness and accuracy. In particular, our analysis reveals that accuracy need not be sacrificed to recover fair graphs. We evaluate FairSpecTemp on synthetic and real-world data sets to illustrate its effectiveness and highlight the advantages of both variants of FairSpecTemp.
Read the full abstract ▾
All authors Madeline Navarro, Andrei Buciulea, Samuel Rey, Antonio G. Marques, Santiago Segarra
Systems and Control
Adaptive Control Allocation for Underactuated Time-Scale Separated Non-Affine Systems
Daniel M. Cherenson and Dimitra Panagou
Many robotic systems are underactuated, meaning not all degrees of freedom can be directly controlled due to lack of actuators, input constraints, or state-dependent actuation. This property, compounded by modeling uncertainties and disturbances, complicates the control design process for trajectory tracking. In this work, we propose an adaptive control architecture for uncertain, nonlinear, underactuated systems with input constraints. Leveraging time-scale separation, we construct a reduced-order model where fast dynamics provide virtual inputs to the slower subsystem and use dynamic control allocation to select the optimal control inputs given the non-affine dynamics. To handle uncertainty, we introduce a state predictor-based adaptive law, and through singular perturbation theory and Lyapunov analysis, we prove stability and bounded tracking of reference trajectories. The proposed method is validated on a VTOL quadplane with nonlinear, state-dependent actuation, demonstrating its utility as a unified controller across various flight regimes, including cruise, landing transition, and hover.
Read the full abstract ▾
Signal Processing
Time-Frequency Filtering Meets Graph Clustering
Marcelo A. Colominas, Stefan Steinerberger and Hau-Tieng Wu
We show that the problem of identifying different signal components from a time-frequency representation can be equivalently phrased as a graph clustering problem: given a graph $G=(V,E)$ one aims to identify `clusters', subgraphs that are strongly connected and have relatively few connections between them. The graph clustering problem is well studied, we show how these ideas can suggest (many) new ways to identify signal components. Numerical experiments illustrate the ideas.
Read the full abstract ▾
Audio and Speech Processing · Cross-listed from cs.CL
Can Speech LLMs Think while Listening?
Yi-Jen Shih, Desh Raj, Chunyang Wu et al. (+6 more)
Recent advances in
speech large
language models (
speech LLMs) have enabled seamless spoken interactions, but these systems still struggle with complex
reasoning tasks. Previously, chain-of-thought (CoT) prompting or fine-tuning has been to shown to significantly improve the
reasoning abilities of text-based LLMs. In this work, we investigate the effect of CoT fine-tuning for multi-stream
speech LLMs, demonstrating that
reasoning in text space improves the accuracy of
speech LLMs by 2.4x, on average, over a suite of spoken
reasoning tasks. Beyond accuracy, the latency of the spoken response is a crucial factor for interacting with voice-based
agents. Inspired by the human behavior of "thinking while listening," we propose methods to reduce the additional latency from
reasoning by allowing the model to start
reasoning before the user query has ended. To achieve this, we introduce an
entropy-based metric, "question completeness," which acts as an indicator to guide the model on the optimal time to start
reasoning. This method provides greater control over the accuracy-latency trade-off compared with heuristic-based approaches and, under equivalent latency conditions, yields a 4% accuracy gain on ARC-Easy. Finally, we use Direct Preference Optimization (DPO) on preference data created using rejection sampling to push the accuracy-latency pareto frontier further, resulting in a 70% reduction in latency without loss in accuracy.
Read the full abstract ▾
All authors Yi-Jen Shih, Desh Raj, Chunyang Wu, Wei Zhou, SK Bong, Yashesh Gaur, Jay Mahadeokar, Ozlem Kalinli, Mike Seltzer
Signal Processing
Flexible Intelligent Metasurface for Reconfiguring Radio Environments
Hanwen Hu, Jiancheng An, Lu Gan et al. (+1 more)
Flexible intelligent metasurface (FIM) technology holds immense potential for increasing the spectral efficiency and energy efficiency of wireless networks. In contrast to traditional rigid reconfigurable intelligent surfaces (RIS), an FIM consists of an array of elements, each capable of independently tuning electromagnetic signals, while flexibly adjusting its position along the direction perpendicular to the surface. In contrast to traditional rigid metasurfaces, FIM is capable of morphing its surface shape to attain better channel conditions. In this paper, we investigate the single-input single-output (SISO) and multiple-input single-output (MISO) communication systems aided by a transmissive FIM. In the SISO scenario, we jointly optimize the FIM phase shift
matrix and surface shape to maximize the end-to-end channel gain. First, we derive the optimal phase-shift
matrix for each tentative FIM surface shape to decompose the high-dimensional non-convex optimization problem into multiple one-dimensional subproblems. Then, we utilize the particle swarm optimization (PSO)
algorithm and the multi-interval gradient descent (MIGD) method for updating the FIM's surface shape to maximize the channel gain. In the MISO scenario, we jointly optimize the transmit beamforming, the FIM surface shape, and the phase shift
matrix to maximize the channel gain. To tackle this complex problem with multiple highly coupled variables, an efficient alternating optimization
algorithm is proposed. Simulation results demonstrate that FIM significantly improves channel gain compared to traditional RIS and exhibits good adaptability to multipath channels.
Read the full abstract ▾
All authors Hanwen Hu, Jiancheng An, Lu Gan, Naofal Al-Dhahir
Audio and Speech Processing · Cross-listed from cs.CL
LASER: An LLM-based ASR Scoring and Evaluation Rubric
Amruta Parulekar and Preethi Jyothi
Standard ASR evaluation metrics like Word Error Rate (WER) tend to unfairly penalize morphological and syntactic nuances that do not significantly alter sentence semantics. We introduce an LLM-based scoring rubric
LASER that leverages state-of-the-art LLMs' in-context learning abilities to learn from
prompts with detailed examples. Hindi
LASER scores using Gemini 2.5 Pro achieved a very high correlation score of 94% with human annotations. Hindi examples in the
prompt were also effective in analyzing errors in other Indian
languages such as Marathi, Kannada and Malayalam. We also demonstrate how a smaller LLM like Llama 3 can be finetuned on word-pair examples derived from reference and ASR predictions to predict what kind of penalty should be applied with close to 89% accuracy.
Read the full abstract ▾
Signal Processing
Autoencoding Coordinate Sequences from Psychophysiologic Signals
Timothy L. Hutcheson and Anil K. Raj
We present a method for converting 24 channels of psychophysiologic time series data collected from individual participants via electroencephalogram (
EEG), electrocardiogram (ECG), electrodermal activity (EDA), respiration rate (RR) into trackable three dimensional (3D) coordinates sufficient to estimate participation in specific task and cognitive states.
Read the full abstract ▾
Audio and Speech Processing
Comparison of Speech Tasks in Human Expert and Machine Detection of Parkinson's Disease
Peter Plantinga, Roozbeh Sattari, Karine Marcotte et al. (+8 more)
The
speech of people with Parkinson's Disease (PD) has been shown to hold important clues about the presence and progression of the disease. We investigate the factors based on which humans experts make judgments of the presence of disease in
speech samples over five different
speech tasks: phonations, sentence repetition, reading, recall, and picture description. We make comparisons by conducting listening tests to determine clinicians accuracy at recognizing signs of PD from audio alone, and we conduct experiments with a machine learning system for detection based on Whisper. Across tasks, Whisper performs on par or better than human experts when only audio is available, especially on challenging but important subgroups of the data: younger patients, mild cases, and female patients. Whisper's ability to recognize acoustic cues in difficult cases complements the
multimodal and contextual strengths of human experts.
Read the full abstract ▾
All authors Peter Plantinga, Roozbeh Sattari, Karine Marcotte, Carla Di Gironimo, Madeleine Sharp, Liziane Bouvier, Maiya Geddes, Ingrid Verduyckt, Étienne de Villers-Sidani, Mirco Ravanelli, Denise Klein
Audio and Speech Processing · Cross-listed from cs.SD
AudioMarathon: A Comprehensive Benchmark for Long-Context Audio Understanding and Efficiency in Audio LLMs
Peize He, Zichen Wen, Yubo Wang et al. (+13 more)
Processing long-form audio is a major challenge for Large Audio
Language models (LALMs). These models struggle with the quadratic cost of
attention ($O(N^2)$) and with modeling long-range temporal dependencies. Existing audio
benchmarks are built mostly from short clips and do not evaluate models in realistic long context settings. To address this gap, we introduce AudioMarathon, a
benchmark designed to evaluate both understanding and inference efficiency on long-form audio. AudioMarathon provides a diverse set of tasks built upon three pillars: long-context audio inputs with durations ranging from 90.0 to 300.0 seconds, which correspond to encoded sequences of 2,250 to 7,500 audio
tokens, respectively, full domain coverage across
speech, sound, and music, and complex
reasoning that requires multi-hop inference. We evaluate state-of-the-art LALMs and observe clear performance drops as audio length grows. We also study acceleration techniques and analyze the trade-offs of
token pruning and KV cache eviction. The results show large gaps across current LALMs and highlight the need for better temporal
reasoning and
memory-efficient architectures. We believe AudioMarathon will drive the audio and
multimodal research community to develop more advanced audio understanding models capable of solving complex audio tasks.
Read the full abstract ▾
All authors Peize He, Zichen Wen, Yubo Wang, Yuxuan Wang, Xiaoqian Liu, Jiajie Huang, Zehui Lei, Zhuangcheng Gu, Xiangqi Jin, Jiabing Yang, Kai Li, Zhifei Liu, Weijia Li, Cunxiang Wang, Conghui He, Linfeng Zhang
Systems and Control · Cross-listed from cs.NI
Anti-Jamming based on Beam-Steering Antennas and Intelligent UAV Swarm Behavior
Tiago Silva and António Grilo
In recent years, Unmanned Aerial Vehicles (
UAVs) have brought a new true revolution to military tactics. While
UAVs already constitute an advantage when operating alone, multi-
UAV swarms expand the available possibilities, allowing the
UAVs to collaborate and support each other as a team to carry out a given task. This entails the capability to exchange information related with situation awareness and action coordination by means of a suitable wireless communication technology. In such scenario, the adversary is expected to disrupt communications by jamming the communication channel. The latter becomes the Achilles heel of the swarm. While anti-jamming techniques constitute a well covered topic in the literature, the use of intelligent swarm behaviors to leverage those techniques is still an open research issue. This paper explores the use of Genetic
Algorithms (GAs) to jointly optimize
UAV swarm formation, beam-steering
antennas and traffic routing in order to mitigate the effect of jamming in the main coordination channel, under the assumption that a more robust and low data rate channel is used for formation management signaling. Simulation results show the effectiveness of proposed approach. However, the significant computational cost paves the way for further research.
Read the full abstract ▾
Image and Video Processing
Content-Adaptive Inference for State-of-the-art Learned Video Compression
Ahmet Bilican, M. Ak{\i}n Y{\i}lmaz and A. Murat Tekalp
While the BD-rate performance of recent learned
video codec models in both low-delay and random-access modes exceed that of respective modes of traditional codecs on average over common
benchmarks, the performance improvements for individual
videos with complex/large motions is much smaller compared to scenes with simple motion. This is related to the inability of a learned encoder model to generalize to motion vector ranges that have not been seen in the training set, which causes loss of performance in both coding of flow fields as well as frame prediction and coding. As a remedy, we propose a generic (model-agnostic) framework to control the scale of motion vectors in a scene during inference (encoding) to approximately match the range of motion vectors in the test and training
videos by adaptively downsampling frames. This results in down-scaled motion vectors enabling: i) better flow estimation; hence, frame prediction and ii) more efficient flow compression. We show that the proposed framework for content-adaptive inference improves the BD-rate performance of already state-of-the-art low-delay
video codec DCVC-FM by up to 41% on individual
videos without any model fine tuning. We present ablation studies to show measures of motion and scene complexity can be used to predict the effectiveness of the proposed framework.
Read the full abstract ▾
Signal Processing
Moments Matter: Posterior Recovery in Poisson Denoising via Log-Networks
Shirin Shoushtari, Edward P. Chandler and Ulugbek S. Kamilov
Poisson denoising plays a central role in
photon-limited imaging applications such as microscopy, astronomy, and medical imaging. It is common to train deep learning models for denoising using the mean-squared error (MSE) loss, which corresponds to computing the posterior mean $\mathbb{E}[x \mid y]$. When the noise is Gaussian, Tweedie's formula enables approximation of the posterior distribution through its higher-order moments. However, this connection no longer holds for Poisson denoising: while $ \mathbb{E}[x \mid y] $ still minimizes MSE, it fails to capture posterior uncertainty. We propose a new strategy for Poisson denoising based on training a log-network. Instead of predicting the posterior mean $ \mathbb{E}[x \mid y] $, the log-network is trained to learn $\mathbb{E}[\log x \mid y]$, leveraging the logarithm as a convenient parameterization for the Poisson distribution. We provide a theoretical proof that the proposed log-network enables recovery of higher-order posterior moments and thus supports posterior approximation. Experiments on simulated data show that our method matches the denoising performance of standard MMSE models while providing access to the posterior.
Read the full abstract ▾
Systems and Control
Stability Preserving Safe Control of a Bicopter
Jhon Manuel Portella Delgado and Ankit Goel
This paper presents a control law for stabilization and trajectory tracking of a multicopter subject to safety constraints. The proposed approach guarantees forward invariance of a prescribed safety set while ensuring smooth tracking performance. Unlike conventional control barrier function methods, the constrained control problem is transformed into an unconstrained one using state-dependent mappings together with carefully constructed Lyapunov functions. This approach enables explicit synthesis of the control law, instead of requiring a solution of constrained optimization at each step. The transformation also enables the controller to enforce safety without sacrificing stability or performance. Simulation results for a polytopic reference trajectory confined within a designated safe region demonstrate the effectiveness of the proposed method.
Read the full abstract ▾
Signal Processing
Towards Reliable Emergency Wireless Communications over SAGINs: A Composite Fading and QoS-Centric Perspective
Yinong Chen, Wenchi Cheng, Jingqing Wang et al. (+2 more)
In emergency wireless communications (EWC) scenarios, ensuring reliable, flexible, and high-rate transmission while simultaneously maintaining seamless coverage and rapid response capabilities presents a critical technical challenge. To this end, satellite-aerial-ground integrated network (SAGIN) has emerged as a promising solution due to its comprehensive three-dimensional coverage and capability to meet stringent, multi-faceted quality-of-service (QoS) requirements. Nevertheless, most existing studies either neglected the inherent characteristics of the complex channel conditions due to the terrain changes or analyzed the performance in the absence of QoS constraints, resulting in a mismatch between theoretical analysis and practical performance. To remedy such deficiencies, in this paper we establish a performance modeling framework for SAGIN employing the Fisher-Snedecor $\mathcal{F}$ composite fading model to characterize the air-ground link. In specific, the proposed $\mathcal{F}$ composite fading channel is adopted to accurately describe both multipath fading and shadowing in harsh ground environments. The exact distribution of end-to-end signal-to-noise (SNR) statistics for space-air and air-ground links is developed, enabling theoretical analysis of cascaded channels with fixed-gain amplify-and-forward (AF) and decode-and-forward (DF) relaying protocols, respectively. Furthermore, asymptotic expressions of the derived results are provided to offer concise representations and demonstrate close
alignment with theoretical predictions in the high-SNR regime. Finally, the insightful closed-form and asymptotic expressions of effective capacity with QoS provisioning, outage probability, and $\epsilon$-outage capacity are investigated, respectively, followed by both field measurements and Monte Carlo simulations to verify the effectiveness.
Read the full abstract ▾
All authors Yinong Chen, Wenchi Cheng, Jingqing Wang, Xiao Zheng, Jiangzhou Wang
Systems and Control · Cross-listed from cs.ET
From Neural Sensing to Stimulation: An Interdisciplinary Roadmap for Neurotechnology
Ruben Ruiz-Mateos Serrano, Joe G Troughton, Nima Mirkhani et al. (+12 more)
Neurotechnologies are transforming how we measure, interpret, and modulate
brain-body interactions, integrating real-time sensing, computation, and stimulation to enable precise physiological control. They hold transformative potential across clinical and non-clinical domains, from treating disorders to enhancing cognition and performance. Realizing this potential requires navigating complex, interdisciplinary challenges spanning neuroscience, materials science, device engineering, signal processing, computational modelling, and regulatory and ethical frameworks. This Perspective presents a strategic roadmap for neurotechnology development, created by early-career researchers, highlighting their role at the intersection of disciplines and their capacity to bridge traditional silos. We identify five cross-cutting trade-offs that constrain progress across functionality, scalability, adaptability, and translatability, and illustrate how technical domains influence their resolution. Rather than a domain-specific review, we focus on shared challenges and strategic opportunities that transcend disciplines. We propose a unified framework for collaborative innovation and education, highlight ethical and regulatory priorities, and outline a timeline for overcoming key bottlenecks. By aligning technical development with translational and societal needs, this roadmap aims to accelerate equitable, effective, and future-ready adaptive neurotechnologies, guiding coordinated efforts across the global research and innovation community.
Read the full abstract ▾
All authors Ruben Ruiz-Mateos Serrano, Joe G Troughton, Nima Mirkhani, Natalia Martinez, Massimo Mariello, Jordan Tsigarides, Simon Williamson, Juan Sapriza, Ioana Susnoschi Luca, Antonio Dominguez-Alfaro, Estelle Cuttaz, Nicole Thompson, Sydney Swedick, Latifah Almulla, Amparo Guemes
Systems and Control
Identification and optimal control strategies for the transversal splitting of ultra–cold Bose gases
Nikolaus Würkner, Yevhenii Kuriatnikov, Karthikeyan Kumaran et al. (+5 more)
Splitting a Bose–Einstein condensate (BEC) is a key operation in fundamental physics experiments and emerging
quantum technologies, where precise preparation of well–defined initial states requires fast yet coherent control of the condensate's nonlinear dynamics. This work formulates the BEC splitting process as an optimal feedforward control problem based on a physically interpretable, reduced–order model identified from limited experimental data. We introduce a systematic calibration strategy that combines optimal experiment selection and constrained nonlinear parameter estimation, enabling accurate system identification with minimal experimental overhead. Using this calibrated model, we compute energy–optimal trajectories via indirect optimal control to realize shortcuts to adiabaticity (STAs), achieving rapid transitions to the ground state of a double–well potential while suppressing excitations. Experiments confirm that the proposed control framework yields high–fidelity state transfers across multiple configurations, demonstrating its robustness and scalability for
quantum control applications.
Read the full abstract ▾
All authors Nikolaus Würkner, Yevhenii Kuriatnikov, Karthikeyan Kumaran, M Venkat Ramana, Jörg Schmiedmayer, Andreas Kugi, Maximilian Prüfer, Andreas Deutschmann-Olek
Systems and Control
Mitigating Increase-Decrease Gaming with Alternative Connection Agreements: A Defender-Attacker-Defender Game
Bart van der Holst, Thomas Swarts, Phuong Nguyen et al. (+2 more)
Redispatch
markets are widely used by system operators to manage network congestion. A well-known drawback, however, is that Flexibility Service Providers (FSPs) may strategically adjust their baselines in anticipation of redispatch actions, thereby aggravating congestion and raising system costs. To address this increase-decrease gaming, Distribution System Operators (DSOs) could use Alternative Connection Agreements (ACAs) to conditionally limit the available connection capacity of
market participants in the day-ahead stage. In this paper, we present a novel Defender-Attacker-Defender game to investigate the potential of this approach in distribution networks under load and price uncertainty. We solve the resulting trilevel optimization model using a custom branch-and-bound
algorithm, and we demonstrate that it efficiently solves the problem without exploring many nodes in the branch-and-bound search tree for most simulated scenarios. The case study demonstrates that applying ACAs can substantially lower redispatch costs (e.g. by 25%) for the DSO with only a limited impact on FSP profits. The effectiveness of the approach critically depends on how often the DSO can invoke ACAs and on the extent to which the DSO can anticipate strategic bidding behavior of the FSP.
Read the full abstract ▾
All authors Bart van der Holst, Thomas Swarts, Phuong Nguyen, Johan Morren, Koen Kok
Audio and Speech Processing · Cross-listed from cs.CL
Modeling Sarcastic Speech: Semantic and Prosodic Cues in a Speech Synthesis Framework
Zhu Li, Yuqing Zhang, Xiyuan Gao et al. (+2 more)
Sarcasm is a pragmatic phenomenon in which speakers convey meanings that diverge from literal content, relying on an interaction between semantics and prosodic expression. However, how these cues jointly contribute to the recognition of sarcasm remains poorly understood. We propose a computational framework that models sarcasm as the integration of semantic interpretation and prosodic realization. Semantic cues are derived from an LLaMA 3 model fine-tuned to capture discourse-level markers of sarcastic intent, while prosodic cues are extracted through semantically aligned utterances drawn from a database of sarcastic
speech, providing prosodic exemplars of sarcastic delivery. Using a
speech synthesis testbed, perceptual evaluations show that semantic and prosodic cues enhance perceived sarcasm, with the combined system achieving the best downstream F1 while maintaining high subjective sarcasm ratings. These findings highlight the complementary roles of semantics and prosody in pragmatic interpretation and illustrate how modeling can shed light on the mechanisms underlying sarcastic communication.
Read the full abstract ▾
All authors Zhu Li, Yuqing Zhang, Xiyuan Gao, Shekhar Nayak, Matt Coler
Systems and Control
Falsification-driven reinforcement learning for maritime motion planning
Marlon Müller, Florian Finkeldei, Hanna Krasowski et al. (+2 more)
Compliance with maritime traffic rules is essential for the safe operation of autonomous vessels, yet training reinforcement learning (RL)
agents to adhere to them is challenging. The behavior of RL
agents is shaped by the training scenarios they encounter, but creating scenarios that capture the complexity of maritime
navigation is non-trivial, and real-world data alone is insufficient. To address this, we propose a falsification-driven RL approach that generates
adversarial training scenarios in which the vessel under test violates maritime traffic rules, which are expressed as signal temporal logic specifications. Our experiments on open-sea
navigation with two vessels demonstrate that the proposed approach provides more relevant training scenarios and achieves more consistent rule compliance.
Read the full abstract ▾
All authors Marlon Müller, Florian Finkeldei, Hanna Krasowski, Murat Arcak, Matthias Althoff
Audio and Speech Processing · Cross-listed from cs.CL
Open ASR Leaderboard: Towards Reproducible and Transparent Multilingual and Long-Form Speech Recognition Evaluation
Vaibhav Srivastav, Steven Zheng, Eric Bezzam et al. (+6 more)
We present the Open ASR Leaderboard, a reproducible benchmarking platform with community contributions from academia and industry. It compares 86 open-source and proprietary systems across 12
datasets, with English short- and long-form and
multilingual short-form tracks. We standardize word error rate (WER) and inverse real-time factor (RTFx) evaluation for consistent accuracy-efficiency comparisons across model architectures and toolkits (e.g., ESPNet, NeMo, SpeechBrain,
Transformers). We observe that Conformer-based encoders paired with
transformer-based decoders achieve the best average WER, while connectionist temporal classification (CTC) and
token-and-duration transducer (TDT) decoders offer superior RTFx, making them better suited for long-form and batched processing. All code and
dataset loaders are open-sourced to support transparent, extensible evaluation. We present our evaluation methodology to facilitate community-driven benchmarking in ASR and other tasks.
Read the full abstract ▾
All authors Vaibhav Srivastav, Steven Zheng, Eric Bezzam, Eustache Le Bihan, Nithin Rao Koluguri, Piotr Żelasko, Somshubra Majumdar, Adel Moumen, Sanchit Gandhi